[R-pkg-devel] Rdmacros as Suggests rather than Imports

James Pustejovsky jepu@to @end|ng |rom gm@||@com
Fri Jan 22 22:42:33 CET 2021


Dirk,

Thanks very much for your input! By "patch it out" do you mean modify my
package description for the Debian distribution only? Would it work to
create a separate branch of my Github repo with modifications for the
Debian distro, and then submit that directly to Debian maintainers?

I'm sorry for my naivete here--I'm just not quite sure how to proceed.

James

On Fri, Jan 22, 2021 at 3:13 PM Dirk Eddelbuettel <edd using debian.org> wrote:

>
> Hi James,
>
> On 22 January 2021 at 14:29, James Pustejovsky wrote:
> | I'm writing to see if anyone has suggestions for how to handle an issue
> I'm
> | facing with the clubSandwich package (
> | https://CRAN.R-project.org/package=clubSandwich; Github repo:
> | https://github.com/jepusto/clubSandwich). The package documentation uses
> | the fairly new package mathjaxr (
> https://CRAN.R-project.org/package=mathjaxr;
> | Github repo: https://github.com/wviechtb/mathjaxr) so that the help
> | packages can include mathjax formulas. Using mathjaxr requires adding it
> as
> | a dependency (in Imports) and in the Rdmacros field.
> |
> | I recently learned that the Debian GNU/Linux team that handles R
> packaging
> | has trouble with the mathjaxr package, and consequently with any package
> | that has mathjaxr as a dependency. Thus, I would like to make
> | mathjaxr an optional feature, moving it from Imports to Suggests.
> | However, Writing R Extensions specifies that RdMacros packages must be
> | listed in Imports or Depends.
> |
> | Does anyone have suggestions for a potential workaround for this
> situation
> | (short of entirely removing the use of mathjaxr)?
>
> That is what Duncan did with rgl, from where this arose. (It's all very
> inside baseball by for embedded 'minified' js files Debian 'ftpmasters'
> [essentially our version of the CRAN desk accept/reject] also require full
> source of the javascript file, which is not unreasonable -- but a bit of a
> pain pain). Wolfgang (for mathjaxr) knows about it, and "time permitting"
> packaging of mathjaxr may change. In the meantime the easiest os for you to
> not use mathjaxr (I don't think the help page rendering mechanism has
> optionality here) or for your corresponding Debian maintainer (CC'ed) to
> patch it out (which is what I had done for the first rgl release with it).
>
> Dirk
>
> --
> https://dirk.eddelbuettel.com | @eddelbuettel | edd using debian.org
>

	[[alternative HTML version deleted]]



More information about the R-package-devel mailing list